class: center, middle, inverse, title-slide # Relationships among urban characteristics, real estate market and spatial patterns of warehouses in different geographic contexts ## Meet xaringan ### true ### true ### 2019/01/15 (updated: 2021-05-09) --- layout: true <div class="my-footer"><span>retaoliveira.github.io/places</span></div> <!-- this adds the link footer to all slides, depends on my-footer class in css--> --- name: xaringan-title class: left, middle background-image: url(img/log7.jpg) background-size: cover .beige[ .larger[ Relationships among urban characteristics, real estate market and spatial patterns of warehouses in different geographic contexts ] ] .beige[ Renata Oliveira | Metting | 10/05/2021 ] <!-- this ends up being the title slide since seal = FALSE--> --- # Authors <img src="img/banner.png" width="90%" style="display: block; margin: auto;" /> .center[ .large[ Renata Oliveira under supervision of Laetita Dablanc and Matthieu Schorung ] ] --- # Research hypotheses .large[ **Logistics sprawl is higher** in cities with a **high differential** between **central and suburban land/rent values**. ] <hr> -- .large[ The **location of warehouses** is closely related to **land/rent values** of logistics facilities. ] --- background-image: url(img/log3.jpg) background-size: cover # Objectives .pull-left[ .left[ **Compare worldwide metropolitan regions considering the relationships among urban attributes, logistics real estate, and logistics facilities' spatial structure focusing on logistics sprawl.** ] ] --- # Methodological approach <img src="img/mermaid.png" width="1539" /> --- # Methodological approach <img src="img/mermaid1.png" width="1539" /> --- # Methodological approach **Spatial descriptive statistics** .pull-left[ .small[ Urban Classification Classification| Category --------------|--------- Lower outlier |Suburban < 95 %| Suburban >= 5 % | Central Upper outlier | Central ] ] .pull-right[ .small[ Warehouse location and rent prices Classification| Category --------------|--------- Lower extreme| Extremely low Lower outlier| Very low < 25 %|Low 25 % - 50 %| Medium 50 % - 75%| Medium > 75 % |High Upper outlier| Very high Upper extreme |Extremely high ] ] --- # Methodological approach **Spatial pattern index** **Univariate Global and Local Moran's I** - spatial autocorrelation. LISA map generated considering a 0.05 level of significance and a Monte Carlo simulation (2000 permutations). --- # Methodological approach **Urban Centrality Index (UCI)*** Computed considering: - **location coefficient** - measure the unequal distribution factor of the urban intensity index within each urban area - **spatial separation index**, namely Venables index, which aims at evaluating the spatial distribution of spatial patterns of activities; and - **proximity index**, which is the normalization of the Venables index considering the respective maximum attainable value. The UCI values range from 0 to 1, where 0 the most polycentric area and 1, maximal monocentricity. .pull-right-narrow[ .small[ *Pereira (2013) ] ] --- # Methodological approach **Dataset containing all hex bins for all the metropolitan regions investigated** .midi[ - Number of warehouses in each hex bin - Average warehouse price in each hex bin - Urban intensity index - Spatial cluster identification for urban intensity index and warehouse count and rent prices - Classification and outlier’s identification for warehouse count and rent prices ] --- # Methodological approach <img src="img/mermaid2.png" width="200%" /> --- # Methodological approach **Dataset containing the summary indicators for each metropolitan region** .midi[ - Global Moran’s I index for urban activity index - Global Moran’s I index for warehouse spatial distribution - Global Moran’s I index for warehouse rent prices distribution - UCI for urban activity index - UCI for warehouse location - Price differential for central and suburban warehouses - Population (t0 and t1) - Metropolitan territorial area - Number of municipalities - Number of warehouses (t0 and t1) - Average distance to barycenter (t0 and t1) - Yearly logistics sprawl - The proportion of HH, HL, LH and LL clusters for the urban activity index, warehouse location and warehouse price. ] --- # Methodological approach **Typology** - K-means cluster analysis .center[ .green[ Can we gather all metropolitan regions into one comparative analysis? ] ] **Dependency** of warehouse location and prices on urban activity - Chi-square independent .center[ .green[ Do warehouse location and prices depend on urban activity? ] ] **Correlation** among continuous variables - Spearman correlation .center[ .green[ Is logistics sprawl related to warehouse rental prices? ] ] --- <img src="img/map.png" width="120%" style="display: block; margin: auto;" />